The elderly are at increased risk of adverse effects from medications, and anecdotal evidence suggests that adverse drug reactions (ADRs) are an important cause of instability, falls and fractures in this vulnerable group. However, there is very little quantitative data available concerning the magnitude of this risk, or comparing the relative risks of specific drugs with one another. We propose to study the relationship between falls and fractures in the elderly and prior use of medications in five commonly prescribed categories: antihypertensives, diuretics, benzodiazepines, neuroleptics, and antidepressants. To accomplish this, the research will employ computer-assisted epidemiological analysis of a large-scale database: the entire Medicaid program of the state of New Jersey for the period 1980-1984 (N=900,000). All health care transactions for all program participants are described in detail in this dataset, comprising some 20 million transactions for recipients over age 35. All patients with any of several diagnostic codes for falls or hip fracture will be identified, as will appropriate controls. Traditional case-control methodologies will then be employed to identify prior medication use patterns and establish relative risk, controlling as needed for confounding factors and effect modifiers. This research will clarify the risk of precipitating gait instability, falls, and associated fractures in elderly patients and will thus lay a foundation of data on which preventive efforts can be built.